Let's build something

I design and build production AI systems — from real-time video pipelines to autonomous trading agents. If you need an engineer who ships infrastructure that actually runs, let's talk.

What I do

AI Video Pipeline Architecture

Multi-model inference chains for live video: object detection, pose estimation, segmentation, and LLM-powered autonomous directing. Built and shipped at Monks on multi-GPU infrastructure.

PyTorchCUDAKafkaKubernetes

Trading System Development

Quantitative research pipelines, walk-forward backtesting, and autonomous trading agents. Built and deployed live systems with real-time inference and full transparency.

PythonPyTorchBacktestingTime-Series

RAG & Agent System Design

Multi-agent orchestration, structured reasoning, and retrieval-augmented generation. Context-aware systems that decompose complex tasks across coordinated sub-agents.

LangChainClaudeVector DBsAgents

ML Infrastructure

Real-time inference serving, GPU pipeline optimization, model deployment, and monitoring. Production ML that handles throughput constraints and hardware limits.

vLLMTensorRTNVIDIA NIMK8s

How I work

Engagement

Hourly or project-based. Scoped deliverables, clear milestones, no fluff.

Remote & async

Timezone-flexible. I communicate through docs, PRs, and short sync calls when needed.

Full stack

From GPU kernels to deployment pipelines. I own the problem end-to-end.

Proof points

  • Media.Monks (S4 Capital) — Lead engineer on real-time AI video production platform with 7+ ML models
  • Banco do Brasil — Enterprise systems for one of Latin America's largest banks
  • Angola National Tax System — Government-scale fiscal infrastructure
  • Curupira — Open-source quant research lab with 31+ strategies tested and published
  • 12+ years building production software across fintech, media, and government

Get in touch

Describe your project and I'll get back to you within 48 hours.